Evaluation of the sensory properties, volatile aroma compounds and functional food potentials of cold-press produced mahaleb ( Prunus mahaleb L.) seed oil

– The aims of this study were to cold-press mahaleb seed, and then fully characterize the oil to extent its food and functional food applications. The novelty of this study relies upon the ﬁ rst data provided by the thermal analysis, sensory analysis, and volatile aroma compounds composition. The seeds were pressed with a screw-type single-head press with a maximum oil exit temperature of 40 ° C. Most common physico-chemical properties, composition analyses, volatile aroma compounds pro ﬁ le, sensory descriptive analysis and consumer tests were completed. The main properties were appropriate and the composition (fatty acids, phytosterols and tocopherols) data concurred with the available literature. Thermal data were provided, and the oil peak crystallization and melting temperatures were (cid:1) 44.45 ° C and (cid:1) 8.41 ° C, respectively. There were 38 volatile aroma compounds quanti ﬁ ed mostly with almond, green, vanillin, woody, and fermented aroma de ﬁ nitions. The panel described the oil with 5 sensory descriptive (almond, vanillin, dough, green, cooling) terms. Consumers liked appearance the most (4.49) with a general acceptance score of 3.70 on a 5-point hedonic scale. Overall, the mahaleb seed oil is a conjugated linolenic, oleic and linoleic fatty acids, b -sitosterol and g -tocopherol rich, very aromatic, and consumer-liked sample. Further studies with various food applications are foreseen.


Introduction
Mahaleb cherry (Prunus mahaleb L.) is a tree classified under the Rosaceae family and Prunoidea subfamily. The tree grows wild and is also cultivated in the Mediterranean, West Asia, and Southeastern Europe regions. It has small and underutilized fruits, but the fruit seeds and seed flours are essential flavoring agents in many baked goods, foods, tonics, and medicines. The seeds have a typical almond-like and vanillin-like flavor, and consequently, its name (mahaleb) comes from the Arabic meaning "sweet aromatic or perfume king" (Özbey et al., 2011;Alma et al., 2012).
This study aimed to characterize cold-press produced mahaleb seed oil and the extent of its possible food and functional food applications. Further, to the best of our knowledge, the thermal properties, volatile aroma compounds composition, sensory descriptive properties, and consumer preferences of the oil were reported for the first time in the literature. Overall, these data provide the novelty of this study.

Materials
The mahaleb seeds were purchased from Bora Baharat Co. (Adana, Türkiye). According to the company statement, the seeds were harvested in November-December 2019 and cultivated in orchards of Tokat Province. The seeds were purchased in June 2020 and stored until the end of the year 2020 in a deep freezer. The seeds were cold-pressed in January 2021, and immediately the analyses were started. All analyses of both replicate oil samples were completed by around June 2021. Consequently, it recommended that the seeds were one year stored samples before cold-pressing. The two replicates of cold-pressing were performed by one weak apart after dividing the seeds into two equal portions at the beginning. The seeds can be observed in Figure 1. All chemicals, solvents, standards, and utensils used were of analytical grade and purchased from Sigma Chem. Co. (St. Louis, US), Merck (Darmstadt, Germany), and local stores.

Seed analyses
The color of the seeds was measured with a Minolta CR-400 Reflectance Colorimeter (Minolta Camera Co., Osaka, Japan) randomly over several points of seed surfaces, and the values of L* (brightness/darkness), a* (redness/greenness) and b* (yellowness/blueness) were recorded. Seed dimensions were measured with a digital caliper (CD-15CP, Mitutoyo Ltd., Andover, UK) to record seed length, width, and thickness. Randomly selected 10 seeds were weighed with a Sartorius ED2245 scale (Sartorius, Germany), and then multiplied by 10 to get 100-seed weight for several times.
After grinding the seeds into fine flour, some compositional analyses were also completed. Total moisture content was determined with an Ohaus MB45 moisture analyzer (Ohaus, Pine Brook, US). Total crude oil content was measured with the Soxhlet technique according to AOAC 920.39 method (AOAC, 2002). Similarly, total crude protein by Kjeldahl according to AOCS method Aa 5-35, and ash content by oven burning following AOCS method Ba 5a-49 were measured (AOCS, 1998).

Cold press oil production
The mahaleb seeds had 5.20% moisture, and before coldpressing, the moisture level of the seeds was adjusted to 10% by the water equilibration process. The amount of added water was calculated by the equation below. The tab water was added to the seeds in a closed vessel and stayed for 24 h at room temperature for moisture penetration and equilibration.
where A is the measured total solids of the seeds (%), B is the targeted total solid of the seeds (%), and C is the total amounts of the seeds (g). After moisture adjustment, the seeds were cold-pressed with a laboratory scale machine (single head, 1.5 kW, 12 kg seed/h capacity; Koçmaksan ESM3710 model, İzmir, Turkey) with 20 rpm screw rotation speed, 14 mm exit die, and 40°C oil exit temperature. Then, the oils were centrifuged (Sigma 2-16 K, Postfach, Germany) at 6797 xg to remove suspended materials and weighed. Finally, the oil was placed into browncolored and capped glass bottles after flushing with nitrogen gas. The oil samples were stored in a refrigerator at 4°C during the analyses. The oil yield of the cold-pressing technique was calculated following the equation below: where A is the total amount of oil obtained by cold pressing (g), and B is the total oil content (g) of the seeds determined by the Soxhlet technique.
The AOCS methods of Ca 5a-40 and Cd 3d-63 were followed to determine oil-free fatty acids (FFA) and acid values (AOCS, 1998). Similarly, oil peroxide value and iodine number were measured by the AOCS Cd 8-53 and Cd 1-25 methods, respectively (AOCS, 1998). The ISO 3596 method was followed to measure oil unsaponifiable matters content (ISO, 1991).

Thermal properties of the oil samples
Melting and crystallization temperatures and enthalpies of the oil samples were assessed with a Perkin-Elmer 4000 Series Differential Scanning Calorimeter-DSC (Groningen, The Netherlands). Around 8 mg sample was weighed and sealed in aluminum pans, and analyzed under 50 mL/min nitrogen flow rate with the following temperature program; cool from 20°C to À60°C by 5°C/min rate, keep oil at that temperature for 3 min, and then heat oil to 60°C by 5°C/min rate. The Pyris 1 manager software was used to calculate the thermal parameters.

Fatty acid, phytosterol and tocopherol composition of the oil samples
The fatty acid composition (%) of mahaleb seed oil was determined by following AOCS methods Ce 2-66 and Ce 1h-05 (AOCS, 1998). An Agilent 7890B gas chromatograph (Agilent Technologies, Palo Alto, CA, USA) equipped with flame ionization detector (FID) (Agilent Technologies, Palo Alto, CA, USA), and HP 88 capillary column (100 m Â 0.25 mmID Â 0.2 mm film thickness, J&W Scientific Co, CA, USA) were used. The thermal oven program was heating at 120°C for 1 min, heating to 175°C (10°C/min) for 10 min, heating to 210°C (5°C/min) for 5 min, and heating to 230°C (5°C/min) for 5 min. The injection volume was 1 mL, the injector split ratio was 1:50, the flow rate was 2 mL/min, hydrogen was the carrier gas, and the injector and the detector temperatures were 250°C and 280°C, respectively. Fatty acid methyl esters were quantified by co-chromatography with FAME mixture standards (37-components, C4-C24, Supelco, Bellefonte, PA, USA). A sample chromatogram can be observed in Figure 2. The peak present at around 28 min was not determined with 37-component standard, but after close comparison with available literature (Özgül-Yücel, 2005;Erdogan Eliuz et al., 2022) providing the peak retention times run in a very similar GC system, it was decided that the peak corresponds to conjugated linolenic acid (CLNA), respectively.
The phytosterol composition (%) of the oils was measured by the ISO 12228 method (ISO, 1999). The unsaponifiable matter was separated previously according to ISO 3596 method (ISO, 1991). After separating the phytosterol fraction from other unsaponifiable matters on TLC, the phytosterol analysis was completed with the same GC and DB5 capillary column (30 m Â 0.25 mm ID Â 0.1 mm film thickness, J&W Scientific Co, CA, USA). The oven program was as follows: keep at 60°C for 2 min, heat to 220°C (40°C/min) for 1 min, and heat to 310°C (5°C/min) for 30 min. The injection volume of 1 mL, injector split ratio of 1:100, and flow rate of 0.8 mL/min, with a hydrogen carrier gas (30 mL/min) was applied. The injector and the detector temperatures were 290°C and 300°C, respectively. Commercially available standards (cholesterol, brassicasterol, stigmasterol, b-sitosterol) were used under the same conditions to identify and quantify the sterols in the samples.
The tocopherol composition (mg/kg oil) of the samples was determined following the method of Grilo et al. (2014) with a reverse-phase HPLC (Shimadzu Corporation, Kyoto, Japan) equipped with an Inertsil ODS-3 column (250 mm Â 4.6 mm Â 5 mm, GL Sciences Inc., Japan), LC-20AT HPLC pump, DGU-20A5R degasser, CTQ-10ASVP column oven, and RF-20A diode array detector. Briefly, 0.15 g oil samples were dissolved in 3 mL dichloromethane, and from this stock, 20 mL were injected by an autosampler (SIL-20AHT) into the HPLC. The methanol: water (98:2 v/v) mobile phase was eluted with a 1.6 mL/min isocratic flow rate. The detector excitation and emission wavelengths were 290 and 330 nm, respectively. Quantification was performed using commercial tocopherol standards (Merck, Darmstadt, Germany).

Volatile aroma compounds composition of the oil samples
Volatile aroma compounds of the oil samples were analyzed following Yılmaz et al. (2015). First, mahaleb seed oil (2 g), salt (0.5 g), and internal standard (20 mL; 1 mL of 2methyl-3-heptanone dissolved in 10 mL methanol) were placed into an amber-colored vial and vortexed for 1 min. Then, the vial was kept at 60°C for 15 min in a water bath before immersing an SPME fiber (2 cm, 50/30 mm DVB/ Carboxen/PDMS, Supelco, Bellafonte) into the vial headspace. The inserted vial was kept for 45 minutes to adsorb the volatiles. Finally, fiber collected volatiles were analyzed with a GC/MS (Agilent 6890HB/Agilent 5875C mass spectrometer; Agilent Technologies, Wilmington, DE, USA) equipped with HP5 MS column (30 m Â 0.25 mm, i.d. 0.25-mm film thickness, J & W Scientific, Folsom, CA, USA). The helium was the carrier gas with a 1.61 mL/min flow rate and 1:2 split ratio. The oven temperature program was waiting at 40°C for 2 min, heating to 200°C (4°C/min), heating to 240°C (7°C/min), and waiting at that temperature for 15 min. The working conditions of MS detector were 280°C capillary direct interface temperature, 70 eV ionization energy, 35-350 amu mass range, 4.45 scans/s scanning rate. A C7-C30 saturated alkane standard was used to calculate the retention indices. Finally, the volatiles tentatively identified using the mass spectra libraries of Wiley, Nist, Tutor, and FFNSC. The data presented as % area values and retention times of the measured volatile compounds.

Sensory analyses of the oil samples
The Quantitative Descriptive Analysis (QDA) method was used to describe the oil samples with a trained panel (Meilgaard et al., 1991). There were 10 panelists (6 female and 4 male, aged between 21-51) and trained for at least 16 h in Panel training was accomplished with oil samples and some food and chemical solution standards to familiarize the panelist with the sensory attribute terms extensively. After training, some pre-tests were completed to measure the panel response ranges, and training continued until all members get trained, and had quite similar test scores. The panel used some commercial cold-pressed oil samples and actual oils produced in this study to determine the sensory terms. The panel defined and described 5 sensory definition terms for this oil. The term "almond" was described as the flavor and aromas associated with almonds, and fresh almonds and benzaldehyde were the standards for 10 point, and absence of the almond was decided with a zero score. "Vanillin" was defined as a sweet-vanilla-like aroma with commercial vanillin as the calibration standard for full score. The term "dough" was defined with the total aromas associated from freshly prepared wheat dough. Fresh wheat flour dough was used as the standard to calibrate the panelist for 10 score. "Green" was defined as aromas associated with cut grass and cis-3hexanol, and cut grass was used as the standard for a total score. Finally, "cooling" was defined as the cool sensation felt in the mouth space after oil tasting, and menthol candy was the standard (10 score) for it. Line scales (10 cm) were used for quantification, anchored with 1 at the left end for minimum intensity and 10 at the right end for maximum intensity. The samples were coded with 3-digit numbers and served in glasses to the panelists with water, unsalted crackers, and expectoration cups. A signed consent form was prepared indicating that the oil sample was food grade and safe, but panelists were suggested not to ingest it during the test. The analyses were conducted under daylight at room temperature in different sessions in a randomized order.
The appearance, smell/aroma, taste/flavor, and general acceptance of the oil samples were evaluated with 171 volunteer consumers (109 men, 62 women, 19-60 years old) using a 5-point hedonic scale (1 = dislike extremely, 5 = like extremely). The samples were coded with 3-digit numbers and served in glasses to the consumers with water, unsalted crackers, and expectoration cups in different sessions. The same consent form was also provided to the consumers (Meilgaard et al., 1991).

Statistical analyses
Cold pressing of the mahaleb seeds was replicated twice. For each replicate sample, the listed analyses were done in triplicate. The collected data were provided as mean values with standard deviations.
3 Results and discussion 3.1 Properties of the mahaleb seeds Some physical properties and proximate composition of the mahaleb seeds used to produce the cold-pressed oils were measured, and the data are summarized in Table 1. The seeds could also be observed in Figure 1. It would be valuable to acquaint the source of the oil. The seeds seem similar to almonds with much smaller dimensions, and their color is brown-yellow with 25.67 b* value and 9.96 a* value. The seed cold-pressed in this study had around 7.73 mm length, 3.43 mm width, and 4.03 mm thickness. The length, width, and thickness of 9.5% and 23% moisture-containing mahaleb seeds were reported as 7.6-8.62 mm, 6.11-7.03 mm and 5.59-7.03 mm, respectively (Yesilo glu and Pınar, 2006). Clearly, seed dimensions changes by seed total moisture level, but the findings concur with literature. The measured 100-seed weight (3.38 g) is in good agreement with the literature reporting a 3.7 g value (Özgül-Yücel, 2005). The measured moisture content (5.20%) indicates that the seeds were dried, and the moisture was within the reported ranges (Mariod et al., 2009;Yesilo glu and Pınar, 2006). Likewise, around 32.12% of total crude oil and 32.65% of crude protein on dry matter basis and 3.87% of ash seem in good agreement with previous studies (Mariod et al., 2009;Alshehri, 2014). Lately, mahaleb seed composition was reported as 4.30-4.69% moisture, 3.21-3.56% ash, 35.01-35.67% fat, 27.52-27.98% protein, and 28.57-29.85% carbohydrate (Dadalı, 2022). The mahaleb seeds used in this study to produce the cold-pressed oil had quite similar composition data to those reported in the literature.

Physico-chemical properties of the cold-pressed mahaleb seed oil
The oil yield was around 24.90% with the cold-pressing of the mahaleb seeds in this study (Tab. 2). This is an expected result for the cold-pressing technique. According to the literature, the cold-pressing technique yield lower quantity oils in expense of higher quality (Yılmaz et al., 2015(Yılmaz et al., , 2020Ok and Yılmaz, 2019). The main physico-chemical properties of the cold-pressed mahaleb seed oil are presented in Table 2. The oil refractive index (1.495) seems to be within the range (1.4-1.5) 4.03 ± 0.39 100-seed weight (gr) 3.38 ± 0.39 defined for vegetable oils and mahaleb seed oil (Mariod et al., 2009;O'Brien, 2004). The specific extinctions K 232 (2.723) and K 270 (0.290) values indicate the level of the primary (hydroperoxides and conjugated dienes) and secondary (carbonyl compounds and conjugated triens) oxidation compounds, and oxidation status of edible oils. The seeds and the oil contain higher quantities of CLNA, and it is known that CLNA is reactive and might interfere with specific extinction measurements, and this situation must be taken into consideration (O'Brien, 2004). During seed storage, some oil oxidation reactions may have also taken place. There is no data in the literature for mahaleb seed oil, but for virgin olive oil, a limit K 232 value of 2.5-2.6, and K 270 value of 0.22-0.25 were defined in the codex standard (Codex, 2017). There is no data about mahaleb seed oil viscosity in the literature, but it was measured as 113.5 cP at 25°C, and this is within the range of similar vegetable oils (O'Brien, 2004). The literature lacks the color values of the mahaleb seed oil, but it seems to be a yellow-greenish oil according to the measured L*, a*, and b* values ( Fig. 1 and Tab. 2). Free fatty acidity of 0.65% oleate and an acid number of 1.31 mg KOH/g were measured (Tab. 2). Mariod et al. (2009) found 1.21-7.86 mg KOH/g acid numbers for the mahaleb seed oil. Turkish food codex for vegetable oils indicates a limit acid value for cold-pressed oils as 4.0 mg KOH/g oil (Codex, 2012). Clearly, our sample is fairly good for these quality parameters. Similarly, the codex limits the peroxide value of cold-pressed oils to a max of 15 meqO 2 /kg oil (Codex, 2012). The measured 4.29 meqO 2 /kg oil peroxide value is far lower than the limit, indicating that the oxidation status of the oil was acceptable. The iodine number of 158.15 gI/100 g oil and unsaponifiable content of 0.72% was measured (Tab. 2). Sbihi et al. (2014) reported Syrian mahaleb seed oil iodine number as 169.2 gI/100 g oil, and unsaponifiable content as 1.3%, and these data concur with this study. Clearly, the mahaleb seed oil is an unsaturated drying oil according to the classification based on iodine numbers (O'Brien, 2004). The chemical quality indices of the cold-pressed mahaleb seed oil were within the ranges reported in the available literature for solvent-extracted mahaleb seed oils, and were within the acceptability limits of the related codex standards. This situation could be accepted as an advantage to exploring the oil as an edible source. Table 3 shows the main fatty acid, phytosterol and tocopherol compositions of the cold-pressed mahaleb seed oil. Sixteen different fatty acids were quantified ( Fig. 2 and Tab. 3), but the majority occurred as CLNA (37.29%), oleic (29.84%), linoleic (26.44%) and palmitic (3.78%) acids. The oil contained a total of 5.42% saturated (SFA), 30.34% monounsaturated (MUFA), and 64.24% of poly-unsaturated fatty acids (PUFA). This data concurs with most of the literature (Özgül-Yücel, 2005;Ercisli and Orhan, 2008;Mariod et al., 2009;Erdogan Eliuz et al., 2022). The major fatty acids of mahaled seed oil were reported as 39.76, 31.33, and 23.01% of CLNA, oleic, and linoleic acids, respectively (Öztürk et al., 2014). In another study (Erdogan Eliuz et al., 2022), 34.39% of CLNA, 31.76% of oleic acid, 25.54% of linoleic acid, and 3.67% of palmitic acid were reported. Sbihi et al. (2015) even reported presence of 40.71% of CLNA in mahaleb seed oil. Clearly, fatty acid data of the present study agrees with literature. The CLNA has been linked with many health benefits including body fat mobilization, cytotoxic activity, and anticancer properties. The edible oils containing CLNA were mostly reported as functional oils (O'Brien, 2004;Ercisli and Orhan 2008;Erdogan Eliuz et al., 2022). CLNA has also classified as reactive fatty acids and prone to oil oxidation, and care during storage and distribution was suggested for those oils containing higher quantities of this fatty acid (O'Brien, 2004).

Fatty acid, phytosterol and tocopherol compositions of the cold-pressed mahaleb seed oil
Eight different phytosterols were quantified (Tab. 3), with the majority of b-sitosterol (88.6%) and campesterol (3.2%), respectively. The major portion of the tocopherols in the mahaleb seed oil (Tab. 3) was g-tocopherol (400.34 mg/kg) and b-tocopherol (348.18 mg/kg), respectively. In one study, g-, a-, and d-tocopherols were quantified in the units of mg/100 g oil as 20.7, 6.4, and 1.4, respectively (Mariod et al., 2009). In another study, 192.52 mg/100 g g-tocopherol, 43.22 mg/100 g d-tocopherol, and 25.84 mg/100 g a-tocopherol were determined (Sbihi et al., 2014). It seems that the tocopherol content in this study is higher than Mariod et al. (2009) and lower than Sbihi et al. (2014). Also, b-tocopherol was present in our sample. These differences could be attributed to the material, processing, and analysis differences.

Thermal properties of the cold-pressed mahaleb seed oil
The DSC-determined thermal properties of the mahaleb seed oil sample are listed in Table 4. Fully melted oil starts crystallization at around À44.45°C, and crystallization completes at around À46.62°C with an enthalpy value of À39.46 J/g. Similarly, mahaleb seed oil's peak melting temperature was determined at À8.41°C (Tab. 4). Since this oil is an oleic-linoleic type oil (Tab. 4), its melting temperature is quite low. It seems to be liquid at refrigerator temperature. There is no thermal data in the literature for mahaleb seed oil, and hence data presented in this study would be important. Sbihi et al. (2014) reported only thermal degradation data for this oil, indicating a degradation temperature of around 350-525°C. Similar thermal behavior for other cold-pressed seed oils in our laboratory was published previously (Yılmaz et al., 2015;Ok and Yılmaz, 2019). The thermal data reported here would be interesting to those who are interested to use this oil.

Volatile aroma compounds composition of the cold-pressed mahaleb seed oil
The volatile aroma compounds quantified in the cold-press produced mahaleb seed oil samples are presented in Table 5. There were 38 different volatile aroma compounds. The highest mean peak area% values were determined for 1-pentanol (9.69%), n-pentanal (9.30%), limonene (3.01%), 2-methylnonane (2.75%), coumarin (2.46%), E-hexenal (2.36%), and 4-methyl nonane (2.15%), respectively. The most common aroma descriptors of these compounds were fermented, almond, fruity, citrus, sweet, hay, terpene, green, and balsamic. Since, in aroma science, the concentration of a compound itself has no meaning, higher concentrations do not mean that they contribute most to the perceived aroma. In fact, the odor (aroma) threshold value indicates how an individual compound contributes to the perceived aroma of a sample. The odor threshold is a minimum concentration value in a specific matrix, indicating that at that lowest concentration, the aroma of the compound could be perceived by human subjects. From the aroma science literature, it is well-known that some compounds produce very intense aroma sensations at low concentrations and some cannot be perceived at even higher concentrations or vice versa. Unfortunately, only a limited number of odor threshold values in specified matrices (aqueous or lipid) are known. Consequently, the perceived characteristic aroma of a food sample is rather a cumulative effect of all volatile aroma compounds present (Meilgaard et al., 1991 ;   Table 3. The fatty acid composition (%), phystosterol composition (%), and tocopherol composition (mg/kg oil) of the cold-pressed mahalep seed oil (Mean ± SD).
As far as we reached, there is no study in the literature on the volatile aroma compounds composition of mahaleb seed oil. The aroma, sugar, and anthocyanin profile of black mahaleb fruit and seed were published (Öztürk et al., 2014).
After comparing the aromatic profiles, it was observed that 3-methyl-1-butanol, n-hexanol, benzyl alcohol, and coumarin were common in their black mahaleb fruit and in our oil sample. Similarly, common volatile aroma compounds of this study (Tab. 5) and their seeds were 1-hexanol, benzyl alcohol and coumarin. In a recent study (Dadalı, 2022), the volatiles of the mahaleb kernel, sour cherry, and apricot kernel were analyzed and compared to determine the adulteration of mahaleb kernel with the other kernels. Mahaleb kernel contained 25 volatiles, including 1 acid, 2 alcohols, 3 aldehydes, 1 furan, 2 ketones, and 16 terpenes. Coumarin was determined as the main volatile, and it was also present in our sample (Tab. 5) with 10 other common volatiles. In another recent study (Dadalı and Elmacı, 2022), the headspace solidphase microextraction technique was optimized, and 21 volatile aroma compounds from terpene, aldehyde, ketone, and alcohol chemical classes were determined from mahaleb kernels. The common volatile aroma compounds in our oil sample and their kernel sample were b-pinene, sabinene, 3-carene, limonene, b-thujene, 2,4-nonadienal, benzyl alcohol, cumin aldehyde, and coumarin. It must be kept in mind that the volatile aroma compounds in this study are from the coldpressed seed oil, while theirs were from fruits and/or seeds (kernels). During seed pressing, cell disruption might have caused some reactions to change the volatiles present. Still, the oil and seed had quite similar distinct volatile aroma compounds. These distinct aromas make this oil special gourmet edible oil. Overall, the volatile aroma compound composition data provided in this study for the cold-press produced mahaleb seed oil could be an important contribution to the literature.

Sensory descriptions of the cold-pressed mahaleb seed oil
The panel defined the mahaleb oil sample with 5 sensory descriptor terms and the mean data generated by the QDA test is summarized in Table 6. Obviously, the most dominant attribute was "almond" with 5.5 mean score on 10 max score indicating the highest intensity. In previous sections, the oil had volatile aroma compounds (Tab. 5) with almond or similar definitions (n-pentanal, 3-methylnonane, coumarin). "Dough" was defined for this oil by the panel with 3.0 score, the second highest value among all. It could also be related with sour and/ or fermented aroma notes. The volatiles (Tab. 5) ethyl alcohol, 3-methyl-1-butanol, 1-pentanol, and 2-butylfuran could be related with the dough descriptor. Similar to most cold-pressed and virgin oils, mahaleb seed oil also had the "green" descriptor (Yılmaz et al., 2015;Ok and Yılmaz, 2019). The volatiles n-butanal, E-2-hexenal, n-hexanol, alpha-thujene, and n-hexanal were defined with green or grassy aroma notes, and these compounds were well perceived by the panel. The panel also defined "vanillin" and "cooling" descriptors with a 1.8 score each. The interaction and combination of the 38 volatile aroma compounds could have created a vanillin like sensation. "Cooling" is the sensation felt in the mouth space, mostly related to mint-like or rapid evaporating compounds. The 2-Methyl-5-isopropenyl-2-cyclohexenone compound was defined as minty licorice could be responsible for the cooling sensation.
There is no sensory study in the literature for mahaleb seed oil. These results could provide important additions to the knowledge about it. In one of our previous studies (Yılmaz et al., 2020), sensory evaluations were determined for the coldpress generated cherry and sour-cherry seed oils. Since the plants are from the same subfamily, there would be some similarities. The sensory definition terms common in both studies were almond and cooling (menthol). Overall, coldpressed mahaleb seed oil was an aromatic-rich oil to be used in various food formulations.

Consumer tests of the cold-pressed mahaleb seed oil
The results of the consumer tests are shown in Table 7. With the 5-point hedonic scale, the highest score for this oil was for appearance (4.49), followed by general acceptance (3.70). Since all attributes (appearance, smell/aroma, taste/ flavor, general acceptance) got scores above the neutrality point (neither like-nor dislike of 3.0 score), it could be claimed that this oil would be accepted by the consumers. Some further processes (pre-roasting, pre-steaming, microwave treatment of seeds, etc.) could be applied to enhance the consumer hedonic scores of the cold-press produced mahaleb oil. Mariod et al. (2009) evaluated solvent-extracted white mahaleb seed oil with 10-point hedonic scale and found 8.7, 8.5, and 8.0 scores for color, odor, and taste, respectively. These findings concur with ours that the appearance and/or color of the oil was liked most.

Conclusions
In this study, cold-press produced mahaleb seed oil was characterized to extend the food and functional food applications of this special oil. Although literature reports some data about solvent-extracted mahaleb seed oils, this study provides the first data about cold-pressed samples. Further, some physico-chemical data (color, viscosity, thermal properties etc.), sensory data, and volatile aroma compounds composition data were reported for the first time in the literature for this oil. The oil has been consumed as an edible oil in some local regions of Turkey, and it is known to be safe and food-grade. Especially, cold-pressing would provide beneficial oil samples as functional foods or functional ingredients for processed foods since they contain many bio- 3.70 ± 0.11 Table 6. Sensory descriptive properties of of the cold-pressed mahalep seed oil (Mean ± SD).
Mahalep oil Almond 5.5 ± 0.5 Vanillin 1.8 ± 0.2 Dough 3.0 ± 0.7 Green 2.2 ± 0.8 Cooling 1.8 ± 0.5 active and volatile compounds intact. It is expected that edible applications of this oil will be extended and/or new studies about food applications will be studied. The oil seems rich in unsaturated CLNA, oleic and linoleic fatty acids, b-sitosterol, and tocopherols. The distinct almond-like and fermented-like vanillin aroma notes would be the main reason to prefer this oil for some food formulations. In addition, the typical aroma and flavor of the oil could be well preferred for some bakery products, confectionaries, and others. Consumers liked this oil, but care (neutral gas atmosphere or vacuum) must be given to control oil oxidation during production, storage, and distribution. More studies about food applications will be anticipated.

Conflicts of interest
The authors declared that they have no conflict of interest in relation to this article.

Funding
No funding was received to assist with the preparation of this manuscript.

Ethical approval
Not applicable.

Data availability
All data generated or analyzed during this study are included in this published article and its supplementary information files.